This revised proposal builds upon three years of investigation into the genetics of morbid obesity in Utah pedigrees. The first three years of the currently funded grant were to identify, recruit and screen first degree relatives of 75 morbidly obese probands defined as being 100 or more pounds over ideal weight and having at least one first degree relative with morbid obesity. 117 morbidly obese probands with familial obesity and 646 of their relatives have been screened. By the start of the proposed continuation, a total of 150 families with two or more morbidly obese first degree relatives will have been screened. These investigators have shown that morbid obesity is very familial (50% of all probands have a first degree relative with morbid obesity) and that offspring of morbidly obese parents have about 2.6 times increased risk of becoming morbidly obese themselves. They have also shown that there is sufficient genetic heterogeneity, temporal trends, and/or environmental confounding in the expression of morbid obesity that traditional, model-dependent genetic segregation and linkage analyses will probably not be successful in identifying the genes related to the development of morbid obesity. White cells have been stored from each of these families in preparation for this continuation proposal. The specific aim of this proposal is to identify and map genes linked to or associated with morbid obesity in Utah families. Fifty of the screened Caucasian pedigrees will be expanded to the cousins of the proband and their offspring (four generations averaging 35 people per pedigree) in order to have 10 or more morbidly obese family members per pedigree. Pilot studies on these pedigrees and long experience with pedigree expansion have indicated this is entirely feasible. Candidate genes suggested in the literature, followed by a general genome search for linked genes, will proceed using markers spanning the genome at an average of 10 cM. With 10 or more affected persons in each of 50 pedigrees, there will be power to detect linkage of a marker to a morbid obesity gene within a single pedigree if certain assumptions are met. This methodology will provide a tremendous advantage by removing inter- pedigree heterogeneity. To detect association with the trait locus, flanking polymorphisms to linked markers will be identified so that the genes may be phenotypically characterized to suggest pathophysiologic mechanisms of action.